IER DEGs

Author

Rasmus Olofzon

Summary and overview

Highlights:

  • UMAPs with sample and cell cycle phase visualized: Figure 1
  • Metadata for dataset: Table 1
  • Cell cycle phase classification distribution per condition: Table 2
  • Previously performed DEG analysis, comparisons shown here explained in this diagram: Figure 2
  • New DEG analysis (explanatory diagram in Figure 6)

Setup

Code 
library(Seurat)
library(stringr)
library(dplyr)
library(ggplot2)
Code 
sobj <- readRDS("../data/processed/seurat_object_w_stress_sig.rds")
Code 
sobj
An object of class Seurat 
32293 features across 12566 samples within 3 assays 
Active assay: RNA (32285 features, 2000 variable features)
 2 other assays present: ADT, HTO
 2 dimensional reductions calculated: pca, umap
Code 
cols_features <- c("moccasin", "darkslategray")
Code 
phases <- list("G1", "G2M", "S")
names(phases) <- phases
deg_files <- lapply(phases, FUN = \(x) str_c("../data/processed/deg/deg_", x, "_ice_t_vs_37c_t.tsv"))
# deg_files
Code 
degs <- lapply(phases, FUN = \(phase) read.table(deg_files[[phase]]))

Visualization of dataset with conditions, for reference

Metadata:

Code 
sobj[[]] %>% head
Table 1: Metadata for dataset
orig.ident nCount_RNA nFeature_RNA nCount_ADT nFeature_ADT nCount_HTO nFeature_HTO percent.mt hto sample ⋯ buffer_treatment incubation_method S.Score G2M.Score Phase old.ident RNA_snn_res.0.8 seurat_clusters stress_signature1 is_stressed
<chr> <dbl> <int> <dbl> <int> <dbl> <int> <dbl> <chr> <chr> ⋯ <chr> <chr> <dbl> <dbl> <chr> <fct> <fct> <fct> <dbl> <chr>
AAACCCAAGAGACAAG-1 DB_AKC_citeseq 11976 3437 128 4 32 3 2.104208 HTO2 37c_no_t ⋯ no_triptolide 37c -0.2740383 -0.19990412 G1 HTO2 0 0 0.19599848 stressed
AAACCCAAGAGTGAAG-1 DB_AKC_citeseq 21028 5475 54 4 108 4 2.425338 HTO3 ice_t ⋯ with_triptolide ice 0.2386460 -0.01205648 S HTO3 1 1 -0.10710847 not_stressed
AAACCCAAGCGAAACC-1 DB_AKC_citeseq 10688 2813 56 3 87 2 4.519087 HTO4 37c_t ⋯ with_triptolide 37c -0.1966232 -0.14132145 G1 HTO4 5 5 -0.04234327 not_stressed
AAACCCAAGGTAAAGG-1 DB_AKC_citeseq 10627 3677 68 4 79 4 2.888868 HTO1 ice_no_t ⋯ no_triptolide ice -0.1555948 -0.21232288 G1 HTO1 2 2 -0.04919840 not_stressed
AAACCCAAGGTCTACT-1 DB_AKC_citeseq 16865 3915 106 4 99 2 2.514082 HTO4 37c_t ⋯ with_triptolide 37c 0.2038750 0.17149954 S HTO4 7 7 -0.08366769 not_stressed
AAACCCAAGTCGGCCT-1 DB_AKC_citeseq 13939 4192 60 3 75 2 3.285745 HTO3 ice_t ⋯ with_triptolide ice -0.3008664 -0.25975436 G1 HTO3 2 2 -0.09381362 not_stressed
Code 
DimPlot(sobj, group.by = "sample") + coord_fixed()
DimPlot(sobj, group.by = "Phase") + coord_fixed()

Figure 1: UMAP coloured on metadata

(a) Coloured on condition/sample

(b) Coloured on cell cycle phases

Code 
table(sobj[[]][,c("sample", "Phase")]) %>% as.data.frame.matrix
Table 2: Cell cycle phase classification distribution per condition
G1 G2M S
<int> <int> <int>
37c_no_t 2164 334 375
37c_t 1799 608 1173
ice_no_t 1825 351 872
ice_t 1760 360 945

Old DEG analysis

From former analysis we performed differential gene expression testing cell cycle phase-wise. The comparison was between conditions/samples. This yielded the following set of files:

Code 
system("ls ../data/processed/deg/", intern = T)
  1. 'Archive.zip'
  2. 'deg_G1_37c_no_t_vs_37c_t.tsv'
  3. 'deg_G1_ice_no_t_vs_37c_no_t.tsv'
  4. 'deg_G1_ice_no_t_vs_ice_t.tsv'
  5. 'deg_G1_ice_t_vs_37c_t.tsv'
  6. 'deg_G1_ice_vs_37c_no_t.tsv'
  7. 'deg_G1_ice_vs_37c_t.tsv'
  8. 'deg_G2M_37c_no_t_vs_37c_t.tsv'
  9. 'deg_G2M_ice_no_t_vs_37c_no_t.tsv'
  10. 'deg_G2M_ice_no_t_vs_ice_t.tsv'
  11. 'deg_G2M_ice_t_vs_37c_t.tsv'
  12. 'deg_G2M_ice_vs_37c_no_t.tsv'
  13. 'deg_G2M_ice_vs_37c_t.tsv'
  14. 'deg_S_37c_no_t_vs_37c_t.tsv'
  15. 'deg_S_ice_no_t_vs_37c_no_t.tsv'
  16. 'deg_S_ice_no_t_vs_ice_t.tsv'
  17. 'deg_S_ice_t_vs_37c_t.tsv'
  18. 'deg_S_ice_vs_37c_no_t.tsv'
  19. 'deg_S_ice_vs_37c_t.tsv'
Figure 2: Schema for (old) DEG comparisons to follow.
flowchart TB
  subgraph ice with triptolide
  G1_ice[G1]
  G2M_ice[G2M]
  S_ice[S]
  end
  subgraph 37c with triptolide
  G1_37c[G1] --- G1_ice
  G2M_37c[G2M] --- G2M_ice
  S_37c[S] --- S_ice
  end

For each file, both up- and down-regulated genes are included. Here we have loaded the comparison between 37c with triptolide and ice with triptolide (see Figure 2). So first we have the top 6 ice-specific, for the G1 phase:

degs[["G1"]] %>% head
A data.frame: 6 × 5
p_val avg_log2FC pct.1 pct.2 p_val_adj
<dbl> <dbl> <dbl> <dbl> <dbl>
Zfp36l2 0 1.958553 0.990 0.439 0
Mir142hg 0 1.894032 0.962 0.245 0
Txnip 0 1.855793 0.945 0.183 0
Sox4 0 1.635571 0.945 0.458 0
Dapp1 0 1.552930 0.966 0.568 0
Gcnt2 0 1.479956 0.959 0.586 0

And the top 6 37c-specific, for the G1 phase as well:

degs[["G1"]] %>% tail
A data.frame: 6 × 5
p_val avg_log2FC pct.1 pct.2 p_val_adj
<dbl> <dbl> <dbl> <dbl> <dbl>
Sh2b2 0.000000e+00 -1.357543 0.093 0.691 0.000000e+00
A930015D03Rik 3.628043e-15 -1.372958 0.487 0.298 1.171314e-10
Asah2 7.332960e-212 -1.375551 0.322 0.723 2.367446e-207
Bcl2l14 3.884711e-151 -1.661921 0.061 0.419 1.254179e-146
Gm40841 0.000000e+00 -1.682257 0.040 0.759 0.000000e+00
4933406J09Rik 0.000000e+00 -2.247251 0.101 0.844 0.000000e+00

DEGs visualized on UMAP

G1

Code 
FeaturePlot(sobj, features = degs[["G1"]] %>% head %>% rownames, coord.fixed = T, order = T, cols = cols_features)
FeaturePlot(sobj, features = degs[["G1"]] %>% tail %>% rownames, coord.fixed = T, order = T, cols = cols_features)

Figure 3: Top DEGs for phase G1.

(a) Top 6 DEGs specific to ice w/ t condition

(b) Top 6 DEGs specific to 37c w/ t condition

G2M

Code 
FeaturePlot(sobj, features = degs[["G2M"]] %>% head %>% rownames, coord.fixed = T, order = T, cols = cols_features)
FeaturePlot(sobj, features = degs[["G2M"]] %>% tail %>% rownames, coord.fixed = T, order = T, cols = cols_features)

Figure 4: Top DEGs for phase G2M.

(a) Top 6 DEGs specific to ice w/ t condition

(b) Top 6 DEGs specific to 37c w/ t condition

S

Code 
FeaturePlot(sobj, features = degs[["S"]] %>% head %>% rownames, coord.fixed = T, order = T, cols = cols_features)
FeaturePlot(sobj, features = degs[["S"]] %>% tail %>% rownames, coord.fixed = T, order = T, cols = cols_features)

Figure 5: Top DEGs for phase S.

(a) Top 6 DEGs specific to ice w/ t condition

(b) Top 6 DEGs specific to 37c w/ t condition

New DEG analysis

Specifically done not cell cycle phase-wise. Will be done for:

  • 37c with tript VS ice with tript
Figure 6: Schema for new DEG comparison.
flowchart TB
  subgraph ice with triptolide
  ice[G1, G2M, S]
  end
  subgraph 37c with triptolide
  37c[G1, G2M, S] --- ice
  end

37c with tript vs ice with tript

Sample names used in object:

Code 
sobj[[]][["sample"]] %>% unique
  1. '37c_no_t'
  2. 'ice_t'
  3. '37c_t'
  4. 'ice_no_t'
Idents(sobj) <- "sample"
DefaultAssay(sobj) <- "RNA"
new_degs <- FindMarkers(sobj, ident.1 = "37c_t", ident.2 = "ice_t", assay = "RNA", features = VariableFeatures(sobj)) %>% arrange(desc(avg_log2FC))

DEG table

Code 
new_degs
Table 3: Differentially expressed genes between 37c_t and ice_t. Positive fold change = up in 37c_t, negative fold change = up in ice_t.
p_val avg_log2FC pct.1 pct.2 p_val_adj
<dbl> <dbl> <dbl> <dbl> <dbl>
4933406J09Rik 0.000000e+00 2.0847480 0.862 0.115 0.000000e+00
Bcl2l14 0.000000e+00 1.6003084 0.465 0.052 0.000000e+00
Gm40841 0.000000e+00 1.5380264 0.771 0.039 0.000000e+00
A930015D03Rik 3.476910e-33 1.2839138 0.339 0.552 1.122520e-28
Sh2b2 0.000000e+00 1.2106941 0.709 0.127 0.000000e+00
Asah2 0.000000e+00 1.2047546 0.730 0.334 0.000000e+00
Arhgap15 0.000000e+00 1.1174486 0.996 0.989 0.000000e+00
Pex1 0.000000e+00 1.0954294 0.599 0.195 0.000000e+00
Gm28403 0.000000e+00 1.0177440 0.566 0.076 0.000000e+00
Mctp1 9.482904e-216 0.9753501 0.951 0.923 3.061555e-211
Ccser1 4.533434e-135 0.9561586 0.773 0.678 1.463619e-130
4930435F18Rik 0.000000e+00 0.8894889 0.494 0.005 0.000000e+00
Diaph3 4.369147e-115 0.8549220 0.699 0.592 1.410579e-110
Diaph2 9.363747e-142 0.8476800 0.736 0.609 3.023086e-137
Cep128 6.111412e-98 0.8312018 0.620 0.509 1.973069e-93
Gm36431 0.000000e+00 0.8201461 0.396 0.012 0.000000e+00
Tmbim7 0.000000e+00 0.8201281 0.425 0.008 0.000000e+00
Gm5099 0.000000e+00 0.7989243 0.513 0.008 0.000000e+00
Pard3b 2.769102e-149 0.7973588 0.875 0.808 8.940047e-145
Kcnq1ot1 1.714648e-114 0.7565798 0.908 0.864 5.535742e-110
Prkg1 4.200709e-63 0.7341202 0.827 0.799 1.356199e-58
Acot12 0.000000e+00 0.7065082 0.455 0.003 0.000000e+00
Fgfr2 4.977226e-110 0.7027180 0.658 0.531 1.606897e-105
Mir99ahg 1.173616e-66 0.7014709 0.816 0.826 3.789018e-62
Rad51b 1.169004e-37 0.6700331 0.572 0.502 3.774131e-33
Nek10 4.344622e-288 0.6620993 0.373 0.011 1.402661e-283
Map3k15 5.950301e-300 0.6444104 0.409 0.026 1.921055e-295
Gna14 7.698322e-271 0.6443333 0.403 0.041 2.485403e-266
Hormad2 1.147156e-251 0.6362092 0.367 0.028 3.703594e-247
Esr1 1.082358e-81 0.6282468 0.715 0.623 3.494394e-77
⋮ ⋮ ⋮ ⋮ ⋮ ⋮
Fam111a 4.671189e-242 -0.7086627 0.335 0.679 1.508093e-237
Runx1 1.348801e-235 -0.7112591 0.813 0.955 4.354603e-231
Dleu2 0.000000e+00 -0.7185027 0.207 0.680 0.000000e+00
Rps6ka5 2.082905e-307 -0.7386294 0.365 0.764 6.724660e-303
Zeb2 1.804149e-182 -0.7529491 0.468 0.754 5.824696e-178
Gm4258 0.000000e+00 -0.7675290 0.222 0.668 0.000000e+00
Ltb 9.087834e-167 -0.7731711 0.309 0.636 2.934007e-162
Tubb4b 1.347083e-245 -0.7773395 0.443 0.801 4.349059e-241
Hist1h1e 3.536736e-154 -0.7796415 0.264 0.619 1.141835e-149
Ikzf2 5.149385e-243 -0.7877484 0.505 0.819 1.662479e-238
Calcrl 2.602565e-237 -0.7918211 0.518 0.816 8.402380e-233
Fut8 2.539314e-214 -0.8042616 0.680 0.908 8.198174e-210
Abhd17b 0.000000e+00 -0.8345477 0.186 0.717 0.000000e+00
Fli1 0.000000e+00 -0.8746406 0.603 0.916 0.000000e+00
Hlf 3.186349e-205 -0.8866952 0.817 0.936 1.028713e-200
Samsn1 0.000000e+00 -0.9064422 0.785 0.970 0.000000e+00
Hist1h1b 1.075679e-12 -0.9673167 0.297 0.386 3.472828e-08
Igkc 1.736173e-01 -0.9861650 0.429 0.457 1.000000e+00
Il12a 0.000000e+00 -1.0254985 0.091 0.550 0.000000e+00
Myc 0.000000e+00 -1.0260883 0.091 0.675 0.000000e+00
Etv6 0.000000e+00 -1.0842947 0.947 0.994 0.000000e+00
Ikzf1 0.000000e+00 -1.0933579 0.591 0.927 0.000000e+00
Angpt1 2.649381e-294 -1.1610093 0.954 0.985 8.553526e-290
Fchsd2 0.000000e+00 -1.2107342 0.522 0.946 0.000000e+00
Meis1 0.000000e+00 -1.3478146 0.690 0.969 0.000000e+00
Adgrl4 0.000000e+00 -1.3949749 0.931 0.984 0.000000e+00
Slc38a2 0.000000e+00 -1.3967915 0.419 0.957 0.000000e+00
Gcnt2 0.000000e+00 -1.4181788 0.590 0.958 0.000000e+00
Dapp1 0.000000e+00 -1.5052702 0.630 0.977 0.000000e+00
Zfp36l2 0.000000e+00 -1.7229168 0.546 0.992 0.000000e+00

DEG visualizations

Code 
FeaturePlot(sobj, features = new_degs %>% head %>% rownames, coord.fixed = T, order = T, cols = cols_features)
FeaturePlot(sobj, features = new_degs %>% tail %>% rownames, coord.fixed = T, order = T, cols = cols_features)

Figure 7: Top DEGs between 37c with triptolide and ice with triptolide.

(a) Top 6 DEGs specific to 37c_t condition

(b) Top 6 DEGs specific to ice_t condition